Skip to content
SPECIFICTION
Get listed

SongGeneration-v2

github.com

WebLinuxSelf-hosted
Visit github.com
Spec score 7.1 #116 of 123 in AI music generators

Spec sheet

Purpose
LeVo 2 (SongGeneration 2) is described as an open-source music foundation model for generating songs.Read 12 Oct 2026 · github.com
Product
SongGeneration 2, also called LeVo 2, is an open-source music foundation model for AI music generation.Read 11 Oct 2026 · github.com
Supported platforms
webRead 22 Sep 2026 · github.com
Free plan
YesRead 22 Sep 2026 · github.com
Security
The README does not state security or compliance certifications or practices.Read 12 Oct 2026 · github.com
Support
The repository directs users to contact the maker through WeChat or QQ using the displayed QR code.Read 12 Oct 2026 · github.com
Audio upload
YesRead 22 Sep 2026 · github.com
Commercial use
NoRead 22 Sep 2026 · github.com
Compute requirements
The v2-large model listing specifies 22G GPU memory without prompt audio and 28G with prompt audio.Read 12 Oct 2026 · github.com
Generation controls
Generation can be guided by lyrics, text tags for musical attributes, and an optional audio prompt.Read 12 Oct 2026 · github.com
Input constraints
The maker says lyrics require structured section labels and specific punctuation, and recommends comma-separated tags rather than full sentences for descriptions.Read 12 Oct 2026 · github.com
Installation
The README gives a local installation path requiring Python 3.8.12 or newer and CUDA 11.8 or newer, and also documents a Docker option.Read 12 Oct 2026 · github.com
Languages
The listed SongGeneration-v2-large model supports Chinese, English, Spanish, Japanese, and other languages.Read 11 Oct 2026 · github.com
Languages and duration
The v2-large model is listed for Chinese, English, Spanish, Japanese, and other languages, with a maximum song length of 4 minutes 30 seconds.Read 12 Oct 2026 · github.com
Local setup
The repository instructions require Python 3.8.12 or newer and CUDA 11.8 or newer, and also describe running the project with Docker.Read 11 Oct 2026 · github.com
Model limits
SongGeneration-v2-large supports songs up to 4 minutes 30 seconds and lists GPU memory requirements of 22G without prompt audio or 28G with it.Read 11 Oct 2026 · github.com
Model size
The maker says SongGeneration-v2-large has 4 billion parameters.Read 12 Oct 2026 · github.com
Output modes
The model can generate songs with vocals and accompaniment, pure music, vocal-only audio, or separate vocal and accompaniment tracks.Read 12 Oct 2026 · github.com
Outputs
It can generate songs with vocals and accompaniment, pure music, vocal-only audio, or separate vocal and accompaniment tracks.Read 11 Oct 2026 · github.com
Prompt audio limit
For audio prompts longer than 10 seconds, only the first 10 seconds are used.Read 12 Oct 2026 · github.com
Prompt controls
Text descriptions can specify gender, genre, emotion, and instruments; an audio prompt uses its first 10 seconds.Read 11 Oct 2026 · github.com
Prompt limitation
The README advises against combining audio prompts and text descriptions when their instructions conflict because generation quality may degrade.Read 11 Oct 2026 · github.com
Quality caveat
The maker says conflicting information in a text description and audio prompt can reduce generation quality, and that prompt choice can affect performance.Read 12 Oct 2026 · github.com
Security and compliance
The repository page does not state security certifications or compliance claims.Read 11 Oct 2026 · github.com
Song generation
It generates songs from lyrics and can use text descriptions or audio prompts to guide the music.Read 11 Oct 2026 · github.com
Stem export
YesRead 22 Sep 2026 · github.com
Vocal generation
YesRead 22 Sep 2026 · github.com
Web demo
The README links to a Hugging Face Space for trying SongGeneration-v2-Fast, and says it can generate a complete song in under one minute with a slight loss in musicality.Read 12 Oct 2026 · github.com

Where Specifiction places it

Alternatives to SongGeneration-v2All alternatives →